Search

Bacterial catalase is important for intracellular survival of the bacteria. This protein of Propionibacterium acnes, one of possible causes of sarcoidosis, induces hypersensitive Th1 immune responses in sarcoidosis patients. We examined catalase expression in cultured P. acnes isolated from 19 sarcoid and 18 control lymph nodes and immunohistochemical localization of the protein in lymph nodes from 43 sarcoidosis and 102 control patients using a novel P. acnes-specific antibody (PAC) that reacts with the catalase protein, together with the previously reported P. acnes-specific PAB and TIG antibodies. High catalase expression of P. acnes cells was found during stationary phase in more isolates from sarcoid than from non-sarcoid lymph nodes and was associated with bacterial survival under H2O2-induced oxidative stress. In many sarcoid and some control lymph nodes, catalase expression was detected at the outer margins of PAB-reactive Hamazaki-Wesenberg (HW) bodies in sinus macrophages, the same location as catalase expression on the surface of cultured P. acnes and the same distribution as bacterial cell membrane-bound lipoteichoic acid in HW bodies. Some or no catalase expression was detected in sarcoid granulomas with PAB reactivity or in clustered paracortical macrophages packed with many PAB-reactive small-round bodies. HW bodies expressing catalase may be persistent P. acnes in sinus macrophages whereas PAB-reactive small-round bodies with undetectable catalase may be activated P. acnes proliferating in paracortical macrophages. Intracellular proliferation of P. acnes in paracortical macrophages may lead to granuloma formation by this commensal bacterium in sarcoidosis patients with Th1 hypersensitivity to certain P. acnes antigens, including catalase.

BackgroundEarly-onset sarcoidosis (EOS) and Blau syndrome (BS) are systemic inflammatory granulomatous diseases without visible pulmonary involvement, and are distinguishable from their sporadic and familial forms. The diseases are characterized by a triad of skin rashes, symmetrical polyarthritis, and recurrent uveitis. The most common morbidity is ocular involvement, which is usually refractory to conventional treatment. A gain-of-function mutation in the nucleotide-binding oligomerization domain-containing protein 2 (NOD2) gene has been demonstrated in this disease; however, little is known about the relationship between the activation ofNOD2and the pathophysiology of EOS/BS. Here we describe EOS/BS with a novel mutation in theNOD2gene, as well as detection ofPropionibacterium acnes(P. acnes) in the granulomatous inflammation.Case presentationAn 8-year-old Japanese girl presented with refractory bilateral granulomatous panuveitis. Although no joint involvement was evident, she exhibited skin lesions on her legs; a skin biopsy revealed granulomatous dermatitis, andP. acneswas detected within the sarcoid granulomas by immunohistochemistry withP. acnes-specific monoclonal (PAB) antibody. Genetic analyses revealed that the patient had aNOD2heterozygous D512V mutation that was novel and not present in either of her parents. The mutantNOD2showed a similar activation pattern to EOS/BS, thus confirming her diagnosis. After starting oral prednisolone treatment, she experienced an anterior vitreous opacity relapse despite gradual prednisolone tapering; oral methotrexate was subsequently administered, and the patient responded positively.ConclusionsWe presented a case of EOS/BS with a novel D512V mutation in theNOD2gene. In refractory granulomatous panuveitis cases without any joint involvement, EOS/BS should be considered as a differential diagnosis; genetic analyses would lead to a definite diagnosis. Moreover, this is the first report ofP. acnesdemonstrated in granulomas of EOS/BS. Since intracellularP. acnesactivates nuclear factor-kappa B in aNOD2-dependent manner, we hypothesized that the mechanism of granuloma formation in EOS/BS may be the result ofNOD2activity in the presence of the ligand muramyl dipeptide, which is a component ofP. acnes.These results indicate that recognition ofP. acnesthrough mutantNOD2is the etiology in this patient with EOS/BS.

The etiology of sarcoidosis is still obscure; however, Mycobacteria and Propionibacterium acnes are considered the most implicated etiological agent for sarcoidosis. To investigate whether P. acnes is an etiological agent for sarcoid uveitis, we analyzed the frequency of P. acnes detected within the biopsied retinas from patients with ocular sarcoidosis by immunohistochemistry with a P. acnes-specific monoclonal antibody (PAB antibody). Eleven patients (12 eyes) with sarcoid uveitis were enrolled in this study. Eight patients with rhegmatogenous retinal detachment, two patients with non-sarcoid uveitis, and two patients with vitreoretinal lymphoma were enrolled as controls. In the sarcoidosis group, granulomas were mainly observed in the inner retinal layer filled with CD4+ cells and CD68+ cells, indicating the Th1 immune response. P. acnes, identified as round bodies that reacted with the PAB antibody, were present in 10/12 samples (83%) from 9/11 patients (82%) with sarcoidosis. These round bodies were scattered within the retinal granulomas mainly in the inner retinal layer. In the control group, no round bodies were detected. Our results suggested that P. acnes could be associated with sarcoid uveitis. We hypothesize that sarcoid granulomas may be formed by a Th1 immune response to P. acnes hematogenously transmitted to the retina.

Propionibacterium acnes is one of the most commonly implicated etiologic agents of sarcoidosis. We previously reported a complete genome sequence of the C1 strain of P. acnes as a clinical isolate from subcutaneous granulomatous inflammatory lesions in a patient with sarcoidosis. In the present study, we initially searched for genetic profiles specific to the C1 strain by core genome analysis and multiple genome alignment with database sequences from 76 and 9 P. acnes strains, respectively. The analysis revealed that the C1 strain was phylogenetically independent and carried an 18.8-kbp transposon sequence unique to the sarcoid isolate. The unique composite transposon comprised a novel insertion sequence and extrinsic genes from bacteria other than P. acnes. Multilocus sequence typing using 24 sarcoid and 36 non-sarcoid isolates revealed a total of 28 sequence types (STs), including ST26, which was most frequently found without specificity for sarcoid isolates. All 13 ST26 isolates exhibited cell-invasiveness and were confirmed to carry the novel insertion sequence and 4 of the 27 extrinsic CDSs in the transposon, with one exception. ST26 of P. acnes with the composite transposon is the most unique strain detected to date and should be further examined as a causative strain of sarcoidosis.

Wave intensity (WI) is a new hemodynamic index, which is defined as (dP/dt)(dU/dt) at any site of the circulation, where dP/dt and dU/dt are the time derivatives of blood pressure and velocity, respectively. Arterial WI in normal subjects has two positive sharp peaks. The first peak occurs during early systole when a forward-traveling compression wave is generated by the left ventricle. The magnitude of this peak increases markedly with an increase in cardiac contractility. The second peak, which occurs towards the end of systole, is caused by generation of a forward-traveling expansion wave by the ability of the left ventricle to actively stop aortic blood flow. The interval between the R wave of the ECG and the first peak of WI (R-1st peak interval) and the interval between the first and second peaks (1st-2nd interval) are approximately equal to the preejection period and left ventricular ejection time, respectively. Using a combined Doppler and echo-tracking system, we obtained carotid arterial WI noninvasively. We examined the characteristics of WI in 11 patients with mitral regurgitation (MR) before and after surgery, and 24 normal volunteers. In the MR group before surgery, the second peak was decreased and the (1st-2nd interval)/(R-R interval) ratio was reduced, compared with the normal group (140 +/- 130 vs 750 +/- 290mmHg m/s3. P0.0083; 20.7% +/- 3.4% vs 26.7% +/- 2.8%, P0.083). There were no significant differences in the first peak between the normal group and the MR group before and after surgery. The second peak in the MR group was increased significantly (P0.016 vs before surgery) to 1,150 +/- 830mmHg m/s3 in the early period after surgery (stage I), and to 1,090 +/- 580mmHgm/s3 in the late period after surgery (stage II). These values did not differ significantly from that of the normal group. At stage I, the (R-1st peak interval)/ (R-R interval) ratio was increased from 13.4% +/- 2.7% to 20.6% +/- 5.6% (P0.016 vs before surgery). At stage II, this ratio decreased to 16.2% +/- 2.8% (P0.016 vs stage I). but was still significantly higher than that before surgery. The (1st-2nd interval)/(R-R interval) ratio increased significantly after surgery (P0.016 vs before surgery) to values (27.0% +/- 4.5% at stage I and 28.9% +/- 2.6% at stage II) which did not differ significantly from that of the normal group. The recovery of the second peak after surgery suggests that the left ventricle had recovered the ability to actively stop aortic blood flow. Wave intensity is useful for analyzing changes in the working condition of the heart.

To understand the pathophysiology of diseased hearts, we devised a new regional myocardial function curve and investigated its properties. Regional work per unit volume of the myocardium (RWM) was calculated by integrating mean wall stress (σ) with respect to the natural logarithm of the reciprocal of wall thickness [ln(1/H)] over a cardiac cycle. Regarding the end-diastolic ln(1/H) as the preload for the region of concern on the assumption that the myocardium is incompressible, we defined the relation between RWM and end-diastolic ln(1/H) as the regional myocardial function curve. In ten mongrel dogs, we measured left ventricular pressure, left ventricular internal diameter, and wall thickness with a catheter-tip micromanometer and ultrasonic dimension gauges during volume loading to obtain the regional myocardial function curve. We examined the sensitivity of the regional myocardial function curve to changes in contractile state (isoproterenol, propranolol) and changes in afterload (pressure loading by a balloon-occlusion catheter). The linear fit to the data points of the regional myocardial function curve under each condition always achieved a very good correlation coefficient (greater than 0.62). Isoproterenol increased the slope of the regional myocardial function curve from 9.7 ± 0.9 (SEM) mJ/cm3 to 14.4 ± 1.0 mJ/cm3 (P < 0.01), with no significant changes in the x-intercept, while propranolol decreased it to 5.6 ± 1.2mJ/cm3 (P < 0.01) with no significant changes in the x-intercept. The multiple linear regression model with constant x-intercept of the regional myocardial function curve also showed a very high correlation coefficient (greater than 0.98) and sensitivity to changes in contractile state in each dog. The changes in afterload within the physiological range (within a range of peak σ of 4.0–12.0kPa) did not change the slope or the x-intercept of the regional myocardial function curve. We conclude that the relation between RWM and end-diastolic ln(1/H) is a regional myocardial function curve of the ventricle and its slope is useful for comparing myocardial contractility among different regions and different pathological states in the same heart, and among ventricles of different sizes. The present method can be easily extended to the clinical setting.